Globally distributed additive manufacturing is a relatively new frontier in the field of product lifecycle management. Designers are independent of additive manufacturing services, often thousands of miles apart. Manufacturing data must be transmitted electronically from designer to manufacturer to realize the benefits of such a system. Unalterable blockchain legers can record transactions between customers, designers, and manufacturers allowing each to trust the other two without needing to be familiar with each other. Although trust can be established, malicious printers or customers still have the incentive to produce unauthorized or pirated parts. To prevent this, machine instructions are encrypted and electronically transmitted to the printing service, where an authorized printer decrypts the data and prints an approved number of parts or products. The encrypted data may include G-Code machine instructions which contain every motion of every motor on a 3D printer. Once these instructions are decrypted, motor drivers send control signals along wires to the printer's stepper motors. The transmission along these wires is no longer encrypted. If the signals along the wires are read, the motion of the motor can be analyzed, and G-Code can be reverse engineered. This thesis demonstrates such a threat through a simulated attack on a G-Code controlled device. A computer running a numeric controller and G-Code interpreter is connected to standard stepper motors. As G-Code commands are delivered, the magnetic field generated by the transmitted signals is read by a Hall Effect sensor. The rapid oscillation of the magnetic field corresponds to the stepper motor control signals which rhythmically move the motor. The oscillating signals are recorded by a high speed analog to digital converter attached to a second computer. The two systems are completely electronically isolated. The recorded signals are saved as a string of voltage data with a matching time stamp. The voltage data is processed through a Matlab script which analyzes the direction the motor spins and the number of steps the motor takes. With these two pieces of data, the G-Code instructions which produced the motion can be recreated. The demonstration shows the exposure of previously encrypted data, allowing for the unauthorized production of parts, revealing a security flaw in a distributed additive manufacturing environment. / Master of Science / Developed at the end of the 20th century, additive manufacturing, sometimes known as 3D printing, is a relatively new method for the production of physical products. Typically, these have been limited to plastics and a small number of metals. Recently, advances in additive manufacturing technology have allowed an increasing number of industrial and consumer products to be produced on demand. A worldwide industry of additive manufacturing has opened up where product designers and 3D printer operators can work together to deliver products to customers faster and more efficiently. Designers and printers may be on opposite sides of the world, but a customer can go to a local printer and order a part designed by an engineer thousands of miles away. The customer receives a part in as little time as it takes to physically produce the object. To achieve this, the printer needs manufacturing information such as object dimensions, material parameters, and machine settings from the designer. The designer risks unauthorized use and the loss of intellectual property if the manufacturing information is exposed. Legal protections on intellectual property only go so far, especially across borders. Technical solutions can help protect valuable IP. In such an industry, essential data may be digitally encrypted for secure transmission around the world. This information may only be read by authorized printers and printing services and is never saved or read by an outside person or computer. The control computers which read the data also control the physical operation of the printer. Most commonly, electric motors are used to move the machine to produce the physical object. These are most often stepper motors which are connected by wires to the controlling computers and move in a predictable rhythmic fashion. The signals transmitted through the wires generate a magnetic field, which can be detected and recorded. The pattern of the magnetic field matches the steps of the motors. Each step can be counted, and the path of the motors can be precisely traced. The path reveals the shape of the object and the encrypted manufacturing instructions used by the printer. This thesis demonstrates the tracking of motors and creation of encrypted machine code in a simulated 3D printing environment, revealing a potential security flaw in a distributed manufacturing system.

Powder bed fusion (PBF) is one of seven technology modalities categorized under the term additive manufacturing (AM). Beyond the advantages of fabricating complex geometries and the "tool-less manufacturing" paradigm common to all types of AM, polymer PBF shows potential for significant industrial relevance through exploiting the technique's characteristic powder-filled bed (a.k.a. build piston) to utilize the full printer volume for batch-style production. Although PBF should be a suitable processing technique for all semi-crystalline polymers, the polyamide family currently occupies around 90% of the commercial market for polymer PBF. This commercial dominance of polyamides is mirrored in the focus of research publications. The lack of chemical variety in published research questions the universality of reported Structure-Property-Process and Process-Structure-Property relationships for PBF. This dissertation presents the findings from identifying Structure-Property-Process relationships critical to fabricate multi-layer parts for poly(phenylene sulfide) (PPS) by PBF towards expanding PBF material selection and evaluating universality of relationship guidelines. PPS is an engineering thermoplastic used for its high strength, rigidity, dielectric properties, and chemical resistance at elevated temperatures. These properties are attributed to PPS' highly crystalline morphology. Its current use in the automotive and aerospace industries, which are early adopters of AM technologies, makes PPS a prime candidate for AM applications. Therefore, the goal of this work is to demonstrate PPS printing by PBF, study its behavior throughout the PBF lifecycle, and abstract general trends in polymer PBF relationships. First, theoretical ranges for print parameter values are determined from properties of an experimental grade PPS powder feedstock. Successful printing of PPS by PBF is demonstrated in a way contrary to published empirical polymer-PBF relationships. Low temperature printing (i.e., bed temperature more than 15 °C lower than polymer peak melting temperature) of PPS successfully fabricated dimensionally accurate parts with reasonable mechanical properties compared against injection molding values. This distinct PPS behavior does not follow empirical guidelines developed for either polyamides or poly(aryl ether ketones). The unique success of low-temperature PBF prompted further investigation into potential benefits of low-temperature printing. Structure-Property-Process relationships were characterized over the course of simulated powder reuse to show that low-temperature printing prolonged the time when PPS powder properties remained in the "printable" range. Significantly re-used PPS powder was shown to be printable when print parameters were adjusted to accommodate structure and property changes. Successful prints from reused powder is uncommon among published reports of PBF printing of high-performance engineering thermoplastics. Observations of a change in molecular architecture through branching and crosslinking during simulated powder reuse motivated investigating if similar reactions occur in printed parts. PPS is commonly used at elevated temperatures in the presence of oxygen, which is the ideal environment for branching and crosslinking. Structural changes manifested in increased glass transition temperature and high temperature storage modulus. The relative change in structure when printed parts were thermo-oxidatively exposed was observed to be significant for parts printed from new powder, but minimal for parts printed from reused powder. This is a result of the structural changes occurring as powder feedstock during reuse over multiple builds. The changing architecture of reused PPS exposed shortcomings with print parameter value selection based solely on polymer thermal properties. Branching and crosslinking reduced crystallinity, resulting in calculated less energy required to melt; however, it also increased melt viscosity. This negative impact on coalescence behavior was not reflected in the methodology for process parameter value determination because current guidelines neglect rheological properties. These observations motivated proposing a method for selecting print settings based on polymer coalescence behavior. Because it is based on coalescence, this method can predict the transition in governing physics from viscous coalescence to bubble diffusion, which is accompanied by a change in the dependence of mechanical properties on laser energy density. Most work in polymer PBF has focused on "printed part triad'" Process-Property relationships. Work presented in this dissertation contributes to the "printability triad'" of Structure-Property-Process relationships and does so using the novel-to-PBF polymer, PPS. Additional polymers must be explored to continue to discern which polymer-manufacturing relationships are universal among all polymers and which are specific to one subset. The observations and connected interpretation to principles of polymer physics add to the body of knowledge for the polymer PBF field. These contributions will help pave the way for investigations into other polymer families and will re-shape the field's normative logic use when answering the question "what makes a polymer printable by PBF?" Understanding the connection between polymer properties and physical stimuli characteristic of PBF manufacturing will result in parts tailored for specific applications and more sustainable manufacturing, thus realizing additive manufacturing's full potential. / Doctor of Philosophy / Powder bed fusion (PBF) is one of seven distinct additive manufacturing (AM, also known as ``3D printing'') technologies. The manufacturing process creates solid, three-dimensional shapes through selectively heating, melting, and fusing together polymer powder particles in a layer-by-layer manner. Currently, organizations are interested in complementing existing manufacturing technology with PBF for one of three general reasons: (1) "complexity is free" PBF has the ability to make shapes that are difficult or expensive to fabricate using other manufacturing technologies. (2) "tool-less manufacturing" PBF only requires a digital design file to fabricate objects. This enables small changes to be easily made via computer-aided design (CAD) programs without the need to invest time and money into tooling (e.g., molds, jigs, fixtures, or other product-specific tools). This enables "mass customized" products (e.g., custom-fit medical devices and implants) to be economically feasible. (3) "material efficiency" AM is attractive as it often generates less waste than subtractive manufacturing techniques like milling. This is particularly a concern for organizations that manufacture parts from expensive, high-performance polymers, such as in the aerospace and medical industries. Despite these benefits, the state of the art for polymer PBF has room for improvement. Specifically, there are many details regarding material behavior during PBF manufacturing that are unknown; any unknown behaviors present challenges to building confidence in production quality. Additionally, approximately 90% of current PBF use is nylon-12 or else another material in the polyamide family of semi-crystalline thermoplastics. This limited selection of commercially available materials compared against other forms of manufacturing contributes to PBF's circular quandary: the manufacturing process physics are not robustly understood because most experimentation and research has been carried out on one family of polymers; however, a wider variety of polymers has not been developed because there is a limited understanding of the process physics. This dissertation presents research toward answering both PBF challenge areas. The first three chapters present investigations into relationships between the properties of a novel, experimental grade poly(phenylene sulfide) (PPS) semi-crystalline thermoplastic polymer powder, the stimuli imposed on this polymer during PBF processing, and the resultant properties of printed parts (i.e., "property-process-property relationships"). The target polymer, poly(phenylene sulfide), is a high-temperature, high-performance polymer that is traditionally melt processed, but has not yet been commercialized for PBF. Prior literature has established mathematical representation for the interaction between manufacturing energy input and the thermal response of the polymer resulting in melting. This framework has been created through studying the polyamide family. Work presented in this dissertation evaluates existing guidelines for PBF process parameter selection using measured thermal behavior of PPS (i.e., a polysulfide, not a polyamide) to predict the range of manufacturing energies affecting geometrically accurate printed parts of high density and strength. In addition, the impact of thermal exposure from repeated PPS powder reuse over the course of multiple PBF prints was evaluated on powder, thermal, and rheological properties identified as critical for PBF printing. Changes to the molecular structure and properties of reused PPS powder were observed to follow different trends than those reported for other materials traditionally used. The effect of thermal exposure on printed parts was also investigated to determine if the observed changes in molecular structure occurring during thermal exposure of the powder would result in changes to mechanical performance properties of printed parts. The importance of rheological flow properties in dictating printed part performance was observed to be a common theme throughout working with PPS. The final chapter presents a novel method for quantitatively predicting particle fusion during PBF and connecting the extent of particle fusion to mechanical properties of printed parts. The presented method is "polymer agnostic" and advances the state of the art in understanding the physics guiding polymer response to stimuli imposed during PBF AM.

Background: Manufacturing sectors are focusing on developing new manufacturing strategies and improving technologies since there has been a decrease in productivity in recent times. This has led to a massive growth in AM but also due to the benefits of implementing AM technologies. However, there are some challenges to overcome with AM implementation.  Purpose: The purpose of this study is to explore the AM implementation to achieve sustainability in manufacturing companies.   Research questions:  ·      What are the challenges in achieving sustainability with AM implementation for manufacturing companies? ·      How can manufacturing companies achieve environmental and economic sustainability with AM implementation through an industrial technology center?  Method: This study is based on a qualitative method with an abductive approach. The theoretical framework has been gathered through a literature review and the empirical data is based on interviews at the case company. The analysis is based on a thematic analysis method.  Conclusion: This thesis concluded that challenges in achieving sustainability with AM implementation are related to inaccuracies of produced parts and components and geometric complications. This was based on design complexities and printer capabilities. The other conclusion made was that there are aspects that contribute to environmental and economic sustainability through AM implementation such as enhancing process efficiency and low setup costs.

This dissertation presents the innovative design and development of PANDORA, a full-sized humanoid robot that stands 1.9 meters tall and weighs 45 kilograms. Its highly configurable structure was created primarily using Additive Manufacturing(AM) techniques. PANDORA is designed to address the limitations of existing humanoid robots, particularly regarding accessibility, cost, and customization for research purposes. The robot features 32 degrees of freedom, enabling it to perform a wide range of human-like motions, such as walking, reaching, and manipulating objects. The development of PANDORA focuses on leveraging the flexibility of AM to create a lightweight, cost-effective, and easily modifiable robotic platform. The dissertation details the iterative design process, which includes the structural components for weight reduction while maintaining the necessary strength and durability for dynamic movements. The lower body of PANDORA incorporates advanced joint configurations and custom-designed linear actuators, initially developed for previous Terrestrial Robotics and Engineering Controls (TREC) Lab robots, such as THOR and ESCHER. The upper body features a cable-driven arm system, which is both lightweight and highly functional, offering eight degrees of freedom per arm. A significant contribution of this work is the development of design heuristics for AM, tailored specifically for the construction of large-scale robotic components. These heuristics were validated through extensive finite element analysis (FEA) and physical testing, ensuring the AM parts could withstand the loads and stresses encountered during operation. The open-source nature of the PANDORA platform, including all design files and documentation, further enhances its value to the research community, providing a robust foundation for future developments in humanoid robotics. / Doctor of Philosophy / This dissertation explores the creation of PANDORA, a life-sized robot designed to move and function similarly to a human. PANDORA is nearly 6 feet tall and weighs about 100 pounds, making it comparable in size to an average adult. What sets PANDORA apart from other robots is how it was made—using 3D printing technology, which allowed for a strong and lightweight structure. The main goal of this project was to develop a robot that researchers and hobbyists could easily build and modify. To achieve this, PANDORA was designed with affordability and accessibility in mind. By using 3D printing, the number of parts needed to build the robot was significantly reduced, making it easier to assemble and less expensive to produce. The robot's design is also open-source, meaning all the plans and details are freely available online, allowing others to build and improve upon this work. PANDORA has joints that mimic many human movements, such as walking and lifting objects. The arms, for instance, are designed to be both lightweight and highly flexible, making the robot capable of performing tasks that require precision and strength. This research demonstrates how advanced 3D printing can be used to create complex, functional robots and aims to push the boundaries of what is possible in robotics by making these technologies more accessible to everyone.

Keen interest in additive manufacturing (AM) of refractory metals such as tungsten has been motivated by the demand for materials capable of enduring extreme temperatures in aerospace and nuclear applications. The aims of this work were to develop alloy compositions for high-temperature applications in the space propulsion and nuclear fusion sectors, and to establish processing windows for these compositions fabricated using laser powder bed fusion additive manufacturing (LPBF-AM). Tungsten (W)-based alloys are well-suited for high working temperatures because of their high melting points, excellent thermal conductivity, low corrosion resistance, and low coefficient of thermal expansion. The integrated computational materials engineering (ICME) approach was implemented to establish the connections among composition-printability-microstructure-properties-performance framework. ThermoCalc-CALPHAD software was used for Scheil-Gulliver solidification simulation (SGSS) of W-based compositions with various alloying element additions. Chromium, vanadium, and niobium were down-selected as suitable alloying elements based on SGSS results. Further, addition of carbon enhanced printability due to eutectic solidification by the formation of various carbides towards the end of solidification leading to crack-free microstructure as well as being vital for control of oxygen. This work demonstrates the successful manufacturing of multiple crack-free W-based alloy components using LPBF-AM, which had a wide range of working temperatures and enhanced mechanical properties.

Additive manufacturing (AM) is a relatively new domain of manufacturing processes that began with its first patent in 1986. Since then, AM processes quickly grew in popularity due to their flexibility, superior efficiency in high mix low volume manufacturing settings, and lower material costs compared to more subtractive processes. Despite its increasing popularity, AM processes remain behind subtractive processes in terms of quality and the speed at which new technologies are integrated. Introducing Industry 4.0 technologies is an excellent opportunity to address the need for quality assurance tools for AM processes. First, the question of how the quality of additively manufactured parts can be increased to match parts created through subtractive processes must be asked. In this dissertation, two machine learning (ML) models are developed and utilized in a federated environment to mimic what one would see in a production setting. The proposed models increase AM part quality by (1) predicting the resulting geometry of an AM process and (2) compensating for geometric errors by altering the initial stereolithography (STL) file before slicing. In addition to performing geometric error prediction and compensation, the models were enhanced to be resilient to changes in geometry by training on segments of a 3D object rather than the whole object. Next, process parameters from fused-filament fabrication (FFF) processes were added to the ML models to add resilience process parameter variance. Lastly, the ML models were deployed in a federated environment created from three FFF 3D printers that collaboratively created a dataset for the ML models. Collectively, these works expand the research area created by AM, federated learning, and error compensation. This proposal addresses research gaps in the current literature by first setting the prediction and compensation resolution of voxel-based ML methods to a static 100 µm, thereby reducing the error associated with each voxel. Secondly, process parameters are introduced to the model, further increasing prediction and compensation accuracy compared to predicting on the geometry alone. Lastly, the models are deployed in a federated AM environment with multiple 3D printers acting as clients to reduce each client's time spent generating data while maintaining model performance. / Doctor of Philosophy / Additive manufacturing (AM) is a relatively new field where parts are created by extruding material to build a product in the desired shape. A key advantage of such a process is that it is more flexible than those subtractive processes, which remove material from a part. On the other hand, parts produced by AM processes generally have lower quality due to the very specific environments necessary to obtain high-quality parts. Because there is an increased desire to make customized parts (high mix) in small amounts (low volume), AM processes are seeing a rise in popularity, but there is still a need to improve the quality of these produced parts. Furthermore, these environments where AM is utilized generally have multiple 3D printers that manufacturers can leverage to create comprehensive datasets for model development. This dissertation uses machine learning (ML) to collect data from AM processes and reduce AM process errors. By comparing the process's input with the process's output, an ML model can estimate the result of the AM process, including potential defects. This dissertation addresses research gaps in current literature by reducing the error associated with converting the input and output 3D objects to voxels, using parameters to the AM process in the ML models, and using the ML models with 3D printers in a networked environment while forbidding sharing private data.

Wire and Arc Additive Manufacturing is a novel Additive Manufacturing technology. As a result, the process for progressing from a solid model to manufacturing code, i.e. the Part Programming process, is undeveloped. In this report the Part Programming process, unique for Wire and Arc Additive Manufacturing, has been investigated to answer three questions; What is the Part Programming process for Wire and Arc Additive Manufacturing? What are the requirements on the Part Programming process? What software can be used for the Part Programming process? With a systematic review of publications on Wire and Arc Additive Manufacturing and related subjects, the steps of the Part Programming process and its requirements have been clarified. The Part Programming process has been used for evaluation of software solutions, resulting in multiple recommendations for implemented usage. Verification of assumptions, made by the systematic review, has been done by physical experiments to give further credibility to the results.

<p>Aeronautical products additively manufactured by Selective Laser Melting (SLM), are known to have fatigue properties which are negatively impacted by porosity defects, microstructural features and residual stresses. Little research is available studying these phenomena with respect to the short fatigue crack growth (FCG) inconsistency problem, the large focus being on the long FCG. This thesis seeks to add useful knowledge to the understanding of the mechanisms for short crack growth variability in SLM manufactured Ti-6Al-4V, with the two variables for the process conditions and build directions investigated. An in-situ FCG investigation using x-ray synchrotron computed micro-tomography (μXSCT) was used to visually observe and quantify the short crack path evolution. Crack growth, deflections and porosity interactions were noted and discussed in relation to microstructure, build layer thickness and build layer orientation. A novel use of in-situ energy dispersive x-ray diffraction (EDD) was able to show the lattice strains evolving as a propagating crack moved through a small region of interest. The results presented show the ability to reliably obtain all six elastic strain tensor components, and interpret useful knowledge from a small region of interest. </p> <p> </p> <p>There are conflicting views in literature with respect to the damage tolerance behavior of as built SLM manufactured Ti-6Al-4V. In the 2018 review by Agius et al., the more prominent studies were considered with Leuders et al. showing the highest long FCG rates for cracks parallel to the build layer and Cain et al. showing cracks propagating through successive build layers as highest [1]–[3]. Cain et al. and Vilaro et al. report significant anisotropy in long FCG for different build orientations whereas Edwards and Ramulu present similar FCG behavior for three different build directions [2]–[5]. Kruth et al. concluded that for optimized build parameters without any (detectable) pores, the building direction does not play a significant role in the fracture toughness results [6]. All of the mentioned literature reported martensitic microstructures and the presence of prior grain structures for as built SLM Ti-6Al-4V.</p> <p> </p> <p>No studies to the authors knowledge have considered the short FCG of SLM manufactured Ti‑6Al‑4V and its implications to the conflicting damage tolerance behaviors reported in literature [1]. In this work small cross-sectional area (1.5 x 1.5 ) samples in two different build conditions of as built SLM Ti-6Al‑4V are studied. The short FCG rate of three different build directions was considered with cracks parallel to the build layers shown to be the most damaging. The microstructure and build layer are shown to be the likely dominant factors in the short FCG rate of as built Ti-6Al-4V. In terms of porosity, little impact to the propagating short crack was seen although there is local elastoplastic behavior around these defects which could cause toughening in the non-optimized build parameter samples tested. The fracture surfaces were examined using a Scanning Electron Microscope (SEM) with the results showing significant differences in the behavior of the two build conditions. From the microindentation hardness testing undertaken, the smooth fracture surface of the optimized sample correlated with a higher Vickers Hardness (VH) result and therefore higher strength. The non-optimized samples had a ‘rough’ fracture surface, a lower VH result and therefore strength. Furthering the knowledge of short FCG in SLM manufactured Ti-6Al-4V will have positive implications to accurately life and therefore certify additive manufactured aeronautical products.</p>

Additive manufacturing using 2 photon polymerization is of great interest as it can create nanostructures with feature sizes much below the diffraction limit. It can be called as true 3D printing as it can fabricate in 3 dimensions by moving the laser spot in any 3D pattern inside the resist. This unique property is attributed to the non-linearity of two photon absorption which makes the polymerization happen only at the focal spot of the laser beam. This method has a wide range of applications such as optics/photonics, metamaterials, metasurfaces, micromachines, microfluidics, tissue engineering and drug delivery.<br>This work focuses on utilizing 2 photon fabrication for creating a metasurface by printing diabolo antenna arrays on a glass substrate and subsequently metallizing it by coating with gold. A femtosecond laser is used along with a galvo-mirror to scan the geometry inside the photoresist to create the antenna. The structure is simulated using ANSYS HFSS to study its properties and optimize the parameters. The calculations show a reflectance dip and zero reflectance for the resonance condition of 4.04 μm. An array of antennas is fabricated using the optimized properties and coated with gold using e-beam evaporation. This array is studied using a fourier transform infrared spectrometer and polarization dependent reflectance dip to 40% is observed at 6.6 μm. The difference might be due to the small errors in fabrication. This method of 3D printing of antenna arrays and metallization by a single step of e-beam evaporation is hence proved as a viable method for creating optical metasurfaces. Areas of future research for perfecting this method include incorporating an autofocusing system, printing more complicated geometries for antennas, and achieving higher resolution using techniques like stimulated emission depletion.

This research describes the process of manufacturing and machining of wire and arc additive manufactured (WAAM) thin wall structures on integrated and non¬integrated WAAM systems. The overall aim of this thesis is to obtain a better understanding of deposition and machining of WAAM wall parts through an integrated system. This research includes the study of the comparison of deposition of WAAM wall structures on different WAAM platforms, namely an Integrated SAM Edgetek grinding machine, an ABB robot and a Friction Stir Welding (FSW) machine. The result shows that WAAM is a robustly transferable technique that can be implemented across a variety of different platforms typically available in industry. For WAAM deposition, a rise in output repeatedly involves high welding travel speed that usually leads to an undesired humping effect. As part of the objectives of this thesis was to study the travel speed limit for humping. The findings from this research show that the travel speed limit falls within a certain region at which humping starts to occur. One of the objectives of this thesis was to study the effect of lubricants during sequential and non-sequential machining/deposition of the WAAM parts. Conventional fluid lubricants and solid lubricants were used. In addition, the effect of cleaning of deposited wall samples with acetone was also studied. A systematic study shows that a significant amount of solid lubricant contamination can be found in the deposited material. Furthermore, the results indicate that even cleaning of the wire and arc additive manufactured surfaces with acetone prior to the weld deposition can affect the microstructure of the deposited material.